Model comparison
Llama-3.3-70B-Instruct vs Llama 4 Maverick
Llama-3.3-70B-Instruct and Llama 4 Maverick score almost the same on the Noometry Index (30.6 vs 30.9), so choose on price, context window or the category you care about most.
Last verified . 34 shared benchmarks.
MetaSummary
- They share 34 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 3 categories and Llama 4 Maverick in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 4 Maverick leads 26.0 to 15.3.
- The biggest single-benchmark swing is MATH Level 5: 41.6% for Llama-3.3-70B-Instruct and 73% for Llama 4 Maverick.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
Side by side
| Llama-3.3-70B-Instruct | Llama 4 Maverick | |
|---|---|---|
| Provider | Meta | Meta |
| Noometry Index | 30.6 | 30.9 |
| Released | 2024-12-06 | 2025-04-05 |
| Weights | Open | Open |
| Context window | 128K | 128K |
| Max output | 4K | 4K |
| Input $ / M tokens | $0.10 | $0.19 |
| Output $ / M tokens | $0.32 | $0.65 |
| Results tracked | 43 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 31.0 (#290), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| SciCode | 26% | 33.1% |
| WeirdML | 14.4% | 24.5% |
| BigCodeBench Instruct | 46.9% | 49.7% |
| LMArena Coding | 1268 | 1302 |
| BigCodeBench Complete | 57.5% | 61.4% |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LiveBench Coding | 36.6% | — |
| ALE-Bench | — | 172.97 |
Agentic & Tool Use Llama 4 Maverick leads
Llama-3.3-70B-Instruct: 25.8 (#105), Llama 4 Maverick: 28.2 (#91)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | 37.3% |
| BALROG | 23% | — |
Reasoning Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 14.1 (#327), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| SimpleBench | 19.9% | 27.7% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1257 | 1281 |
| DTBench | 59.5% | 61.9% |
| LMCA | 17.5% | 15.9% |
| Epoch Capabilities Index | 127.33 | 132.2 |
| ForecastBench | 58.6 | 57.5 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 55.9% |
| NYT Connections (extended) | — | 8% |
| ARC-AGI-1 | — | 4.4% |
| EnigmaEval | — | 0.6% |
| LiveBench Reasoning | 50.8% | — |
| LiveBench Data Analysis | 49.5% | — |
| LiveBench | 50.2% | — |
Math Llama 4 Maverick leads
Llama-3.3-70B-Instruct: 15.3 (#298), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.1% | 20.6% |
| LMArena Math | 1267 | 1299 |
| MATH Level 5 | 41.6% | 73% |
| Omni-MATH | — | 42.2% |
| LiveBench Math | 42.2% | — |
| FrontierMath (Feb 2025 set) | — | 0.7% |
Knowledge Llama 4 Maverick leads
Llama-3.3-70B-Instruct: 30.6 (#226), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 47.4% | 67% |
| Confabulations | 22.8% | 22.6% |
| Vectara Hallucination Rate | 4.1% | 8.2% |
| LMArena Expert | 1225 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| MMLU-Pro | — | 81% |
| GPQA (HELM) | — | 65% |
| MMLU | 86.3% | — |
Multimodal Not comparable
Llama-3.3-70B-Instruct: —, Llama 4 Maverick: 31.6 (#105)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | — | 1142 |
| GeoBench | — | 52% |
| SpatialViz-Bench | — | 31.8% |
Multilingual Llama 4 Maverick leads
Llama-3.3-70B-Instruct: 39.9 (#220), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1236 | 1269 |
| LMArena Chinese | 1217 | 1277 |
| LMArena French | 1281 | 1259 |
| LMArena German | 1251 | 1291 |
| LMArena Japanese | 1150 | 1207 |
| LMArena Korean | 1143 | 1203 |
| LMArena Russian | 1252 | 1286 |
| LMArena Spanish | 1270 | 1293 |
Instruction Following Too close to call
Llama-3.3-70B-Instruct: 71.1 (#157), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1242 | 1267 |
| LiveBench Instruction Following | 82.7% | — |
| IFEval | — | 90.8% |
Long Context Llama 4 Maverick leads
Llama-3.3-70B-Instruct: 26.4 (#295), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| Fiction.LiveBench | 33.3% | 46.2% |
| LMArena Longer Query | 1256 | 1280 |
Writing & Preference Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 47.6 (#207), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Llama-3.3-70B-Instruct | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1274 | 1287 |
| LMArena Creative Writing | 1250 | 1267 |
| LMArena Multi-Turn | 1280 | 1289 |
| Short-Story Creative Writing | — | 62% |
| EQ-Bench Creative Writing | — | 860 |
| WildBench | — | 80% |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Llama 4 Maverick?
Llama-3.3-70B-Instruct and Llama 4 Maverick score almost the same on the Noometry Index (30.6 vs 30.9), so choose on price, context window or the category you care about most.
Which is cheaper, Llama-3.3-70B-Instruct or Llama 4 Maverick?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.
Is Llama-3.3-70B-Instruct or Llama 4 Maverick better for coding?
Llama-3.3-70B-Instruct scores higher on coding benchmarks: 31.0 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Both accept 128K tokens.
How many benchmarks do Llama-3.3-70B-Instruct and Llama 4 Maverick share?
34 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Llama 4 Maverick has 54.